OBJECTIVE To develop recommendations for the preferred delivery method for a pregnant woman who underwent a previous low transverse cesarean section and who has no contraindications to labor. METHODS MEDLINE searches and the references from retrieved articles yielded 759 citations. Search terms included trial of labor (TOL), trial of scar, vaginal birth after cesarean section (VBAC), and uterine rupture. METHODS Articles with primary outcomes data contrasting TOL and elective repeat cesarean section (ERCS) were analyzed. Studies from developing countries or before 1980 were excluded. METHODS Data from 292 articles were extracted independently by at least 2 team members using a structured form. RESULTS Outcome data were combined using commercially available software and are presented as absolute differences per 10,000, with 95% Bayesian confidence intervals. Maternal outcomes showed that TOL increased the risk for uterine rupture (23.9 [0.24%]); ERCS increased the risk for infection (522 [5.22%]) and bleeding (58.6 [0.59%]). Infant outcomes differed only for 5-minute Apgar scores of less than 7, which were more likely for infants whose mothers underwent TOL (85 [0.85%]). Other outcomes (eg, disability), patient preferences, and cost data did not lend themselves to meta-analysis and were examined separately. While two thirds of women desired TOL, one third preferred ERCS. Costs were 1.7 to 2.4 times greater for ERCS. CONCLUSIONS A woman should be given information on both delivery methods and encouraged to undergo TOL, but her preference for ERCS should be respected.